338 CRIMINAL REPORTS 99 C.R. (6th)
Angelis: Inductive Reasoning, Post-Offence Conduct and Intimate Femicide David M. Tanovich*
Every week in Canada, a woman is killed by a current or former intimate partner.1 It is a serious systemic problem. To put it in perspective, the number of women killed by their intimate partners in 2011 was roughly comparable to the number of gang-related homicides.2 Many, if not most, of these cases involve intimate femicide, a term used to give effect to the gendered nature of the crime. As Rosemary Gartner, Myrna Daw- son and Maria Crawford observe, “. . . intimate femicide is a phenome- non distinct in important ways both from the killing of men by their inti- mate partners and from non-lethal violence against women; and, hence, . . . it requires analysis in its own right.”3 They further observe that: . . . these killings reflect important dimensions of sexual stratifica- tion, such as power differences in intimate relations and the construc- tion of women as sexual objects generally, and as sexual property in
*Faculty of Law, University of Windsor 1See Isabel Grant, “Intimate Femicide: A Study of Sentencing Trends For Men Who Kill Their Intimate Partners” (2010), 47 Alta L Rev 779 at 779 (“[a]pproximately 60 women in Canada are killed each year by their intimate (or former intimate) partners”) (emphasis added) [Grant, “Intimate Femicide”]. See further, Joanne Birenbaum and Isabel Grant, “Taking Threats Seriously: Section 264.1 and Threats as a Form of Domestic Violence” (2012), 59 CLQ 206 at 206–207 (“[in] Canada, a woman is killed by her intimate partner or former intimate partner every six days”) (emphasis added). 2There were 76 women killed by their current or former intimate partners and 95 gang-related homicides in 2011. See Statistics Canada, “Homicide in Canada, 2011” The Daily (December 4, 2012), online:
particular contexts. Intimate femicide — indeed, probably most femi- cide — is not simply violence against a person who happens to be female. It is violence that occurs and takes particular forms because its target is a woman, a woman who has been intimately involved with her killer.4 Other researchers have pointed out that some of the contributing factors that lead to intimate femicide include “possessiveness, . . . the husband accusing the wife of sexual infidelity, . . . her decision to end the rela- tionship, and/or by his desire to control her . . .”5 It is very likely that R. v. Angelis6 is a case of intimate femicide. The most cogent indicators were that the accused had discovered that his wife had been having a long-term affair and that she wanted out of the mar- riage to be with him.7 During a violent confrontation, the accused called his wife a “bitch” and then caused her death. The evidence suggested that she likely asphyxiated either as a result of the accused putting his hand over his wife’s mouth (as described by their eight-year-old daughter) or sitting on top of her until she stopped breathing.8 The jury rejected his claim of self-defence and convicted the accused of second degree mur- der. The Court of Appeal ordered a new trial. The court held that the trial judge had erred in not leaving provocation with the jury despite the fact that the accused disavowed reliance on it. The Court of Appeal also held that the trial judge erred in instructing the jury that it could infer intent from Angelis’ failure to attempt to save his wife once he discovered she was unconscious. It is the latter issue which is the focus of this comment. It begins with a discussion of the process of inductive reasoning which is used to assess the probative value of post-offence conduct in any given case. It then considers the Court of Appeal’s treatment of the accused’s post-offence conduct taking into account the leading precedent of R. v. White.9 The
4Ibid. at 166. 5See Geris Serran & Philip Firestone, “Intimate Partner Homicide: A Review of the Male Proprietariness and Self-Defense Theories” (2004), 9 Aggression and Violent Behavior 1 at 12. 62013 ONCA 70 (Ont. C.A.), reported above at p. 315 [Angelis]. 7Ibid., at para. 9. See further the discussion below at notes 39–40. 8Angelis, supra note 6 at para. 20. 9[2011] 1 S.C.R. 433, 82 C.R. (6th) 11 (S.C.C.) [White]. 340 CRIMINAL REPORTS 99 C.R. (6th)
piece concludes with a consideration of what the common law has taught us about the indicators of intimate femicide and how that was relevant in engaging in inductive reasoning in this case.
The Nature of Inductive Reasoning Although rarely articulated or critically assessed, the law of evidence re- lies on the process of inductive reasoning as the fuel that runs its engine. When, for example, we determine the relevance and probative value of evidence, draw inferences from circumstantial evidence,10 or assess credibility,11 there is often an inferential gap that needs to be filled in order to rationally permit the decision maker or fact finder to do its job.12 To fill that inferential gap we often look to logic, common sense and experience to generate generalizations about human behaviour.13 We then draw a relevant conclusion based on the proven facts and the gener- alization. This process is known as inductive reasoning. In R. v. Munoz,14 one of the few decisions to explore inductive reason- ing, Justice Ducharme described the process as follows: While the jurisprudence is replete with references to the drawing of “reasonable inferences”, there is comparatively little discussion about the process involved in drawing inferences from accepted facts. It must be emphasized that this does not involve deductive reasoning which, assuming the premises are accepted, necessarily results in a valid conclusion. This is because the conclusion is inherent in the relationship between the premises. Rather the process of inference drawing involves inductive reasoning which derives conclusions based on the uniformity of prior human experience. The conclusion is not inherent in the offered evidence, or premises, but flows from
10See, for example, R. v. Quan, 2011 ONCJ 194 (Ont. C.J.). 11See, for example, R. v. Batte (2000), 34 C.R. (5th) 197, 145 C.C.C. (3d) 449 (Ont. C.A.) where Justice Doherty held (at para. 120) that “[j]uries are told to use their common sense and combined life experience in assessing credibility.” 12For example, as noted in R. v. Arcuri, [2001] 2 S.C.R. 828, 44 C.R. (5th) 213 (S.C.C.), “[w]ith circumstantial evidence, there is, by definition, an inferential gap between the evidence and the matter to be established . . .” (at para. 23). 13See the discussion in Hill, Tanovich & Strezos, McWiliams Canadian Crimi- nal Law, 4th ed. (Toronto: Thomson Reuters, 2012) at 28-43–28-49. 14(2006), 38 C.R. (6th) 376, 205 C.C.C. (3d) 70 (Ont. S.C.J.). Angelis: Inductive Reasoning, Post-Offence Conduct and Intimate Femicide 341
an interpretation of that evidence derived from experience. Conse- quently, an inductive conclusion necessarily lacks the same degree of inescapable validity as a deductive conclusion. Therefore, if the premises, or the primary facts, are accepted, the inductive conclusion follows with some degree of probability, but not of necessity.15 Similarly, in R. v. McNair,16 it was further observed that: An inference involves the formation of a conclusion either from in- duction or deduction. In deductive logic an argument is valid if it is impossible for the premises to be true and the conclusion false. . . . Deductive reasoning is “closed”. A conclusion is either valid or not valid. There is no room for compromise in Mr. Spock’s cold deduc- tive Vulcan logic. Inductive reasoning . . . relies for its operation on a level of apprecia- tion and understanding of the paradox inherent in the predictability and unpredictability of events. It goes from what is known to form conclusions about the unknown. The premises of the argument show some degree of inductive probability toward the conclusion but they do not entail the conclusion as in a deductive argument. Inductions are open. There are many conclusions that can reasonably be deter- mined from the same premises...... Using [inductive reasoning] . . . reasonable people may reach different conclusions. . . . There are no precise rules setting out what may be inferred by the process of induction from something else. Shared knowledge, experience or common sense come into play. Sometimes it “makes sense” to conclude with a practical degree of probability that a conclusion follows from certain premises. Some- times the distance from premises to conclusion is so great the degree of probability is minimal. The process of going from premises to conclusion, in this context at least, should not be based on intuition but on factors that can be articulated and related reasonably to the conclusion.17 The challenge for the adversarial process is to enhance the accuracy of inductive reasoning as much as is reasonably possible. This is why social context evidence is so important. Understanding the relevant social con- text ensures that the generalizations relied upon are reasonably reliable
15Ibid. at para. 23 (emphasis added). 162009 CarswellNS 363 (N.S. Prov. Ct.). 17Ibid. at paras. 15–18 (emphasis added). 342 CRIMINAL REPORTS 99 C.R. (6th)
and not infected with stereotypical assumptions.18 As Justice Doherty observed in Peart v. Peel (Regional Municipality) Police Services Board,19 “. . . what makes an appreciation of social context so important [is that] . . . [a]n understanding of how others legitimately view the cir- cumstances serves to counteract the subjectivity of the judge’s own view of the world.”20 An additional challenge is determining when the drivers of inductive reasoning (i.e. logic, common sense and experience) and the social context that informs the drivers require the taking of judicial no- tice or the calling of expert evidence. This difficult and controversial is- sue is beyond the scope of this short note. Here the focus is on the use of social context derived from the cases that have regularly appeared before our appellate courts to inform the drivers of inductive reasoning in determining the admissibility of evidence. In many respects, this is analogous to a court relying on indicators of a so- cial phenomenon to assist in deciding whether it occurred in the particu- lar case. In Peart, for example, the issue was whether the plaintiffs had been the victims of racial profiling. Justice Doherty, for the Court, held: Racial profiling can seldom be proved by direct evidence. Rather, it must be inferred from the circumstances surrounding the police ac- tion that is said to be the product of racial profiling. The courts, as- sisted by various studies, academic writings, and expert evidence have come to recognize a variety of factual indicators that can sup- port the inference that the police conduct was racially motivated, de- spite the existence of an apparent justification for that conduct: R. v. Brown, supra, at paras. 44–46. The indicators of racial profiling recognized in the literature by ex- perts and in the case law can assist a trier of fact in deciding what inferences should or should not be drawn and what testimony should or should not be accepted in a particular case. Those indicators, sometimes referred to as “social” facts, however, cannot dictate the findings that a trier of fact will make in any given case. Findings of adjudicative facts, that is the “who”, “what”, “why”, “when”, and “where” of any given case, grow out of the trier of fact’s assessment of the evidence adduced in the particular case. Findings of adjudica-
18See David M Tanovich, “Relevance, Social Context and Poverty” (2003), 9 C.R. (6th) 348. 19(2006), 43 C.R. (6th) 175 (Ont. C.A.) [Peart]. 20Ibid. at para. 55. Angelis: Inductive Reasoning, Post-Offence Conduct and Intimate Femicide 343
tive facts cannot be preordained by evidence that is intended to pro- vide the appropriate social context in which to assess the evidence and make findings of the relevant adjudicative facts: see R. v. Spence (2005), 202 C.C.C. (3d) 1 at paras. 56–58 (S.C.C.).21
Inductive Reasoning and Post-Offence Conduct Assessing the relevance of the conduct of an accused after an offence involves inductive reasoning. White is the Supreme Court of Canada’s most recent and thorough treatment of the issue. In White, the live issue was whether the accused was guilty of manslaughter, which he conceded, or murder in the shooting of the deceased. The accused testified that his gun went off accidentally during a confrontation. The post-offence con- duct was the accused’s failure to hesitate immediately after firing the gun and before fleeing the scene. The debate before the Supreme Court was whether the inference of intent from that behaviour was reasonable. In his dissenting opinion, Justice Binnie (McLachlin C.J. and Fish J. concurring) concluded that it was not.22 For him, evidence of a lack of hesitation was essentially evidence of demeanour which, like evidence of flight, is highly equivocal and “fraught with danger.”23 In reaching this conclusion, Justice Binnie used judicial experience to inform his inductive reasoning: . . . trial judges spend more time of their working life in and around courtrooms than people summonsed from work or home to jury duty, and experience has taught the judiciary that in some cases jurors have found certain types of evidence (e.g. eyewitness identification and jail-house confessions) more persuasive than was warranted. Misuses
21Ibid. at paras. 95–96 (emphasis added). 22While Justice Binnie dissented on the application of the law to the facts of the case, his opinion is generally regarded as the majority position on the law as it relates to post-offence conduct. In her opinion, Justice Charron (Deschamps J. concurring) stated that she was in “substantial agreement” with Justice Binnie on the law. See White, supra note 9 at paras. 104–107. See further, Lisa Dufraimont, Annotation of R. v. White (2011), 82 C.R. (6th) 14. 23White, supra note 9 at para. 141. Justice Binnie further observed (at para. 142) that “[s]uch demeanour evidence relies too heavily on the witnesses’ power of observation and interpretation, and will often involve a series of speculative in- ferences from a failure to perform as the onlooker thinks ‘normal’ to a conclu- sion of guilt of a particular offence.” 344 CRIMINAL REPORTS 99 C.R. (6th)
of such evidence has on occasion resulted in wrongful convictions. This risk exists with respect to some types of post-offence conduct relied upon by the Crown and in those cases it only makes sense for the judges to alert the jurors to what the courts have collectively learned over the years, especially when the learning may for some jurors be counter-intuitive. . . . The courts have long recognized (though jurors may have no rea- son to know this unless they are told) that the subjective interpreta- tion placed by a witness on the post-offence demeanour evinced by an accused is fraught with danger. These dangers were annotated and persuasively explained in The Commission on Proceedings Involving Guy Paul Morin: Report ...24 Justice Rothstein (LeBel, Abella, Cromwell JJ. concurring) disagreed and held that the evidence was probative of intent: As a matter of logic and human experience, one would expect an ordinary person to present some physical manifestation, such as hesi- tation, at a gun in their hand accidentally discharging into someone’s chest, thereby killing them. It was open to the jury to infer that a failure to react in this way was incongruous with the theory that the gun went off by accident as the two men struggled with each other. ... It is true that the Crown’s use of evidence of lack of hesitation prior to flight presupposes a normal range of reactions: it assumes that most people will hesitate or show some other outward sign of sur- prise when something dramatic and horrible accidentally happens. Of course, it may be that not everyone will respond in this way. How- ever, I consider this view of a normal almost reflexive or involuntary response to be well-founded. . . . Divergence from this norm, though not determinative, is more consistent with an intentional shooting than with an accident.25 In engaging in inductive reasoning, it is troubling that Justice Rothstein did not ground his view of human experience in any expert evidence, social science literature or judicial experience.
24Ibid. at paras. 138, 141. See further, R. v. Biniaris, [2000] 1 S.C.R. 381, 32 C.R. (5th) 1 (S.C.C.) at paras. 39–41, where the Supreme Court set out a judicial experience lens test to assess the reasonableness of verdicts on appeal. 25White, supra note 9 at paras. 67, 70, 79. Angelis: Inductive Reasoning, Post-Offence Conduct and Intimate Femicide 345
Justice Charron (Deschamps J. concurring) agreed with Justice Rothstein that an inference of intent from evidence of a lack of hesitation was not unreasonable or speculative: . . . if there had been significant evidence about the look on the shooter’s face immediately after the fatal shot was fired, I would see no basis for excluding it from the jury’s consideration.26 Six of the nine justices in White thus concluded that evidence of whether an accused hesitates or not after firing a gun and before fleeing the scene can be used on the issue of intent even where the accused admits the actus reus of the offence. As noted earlier, one of the central issues in Angelis involved the admis- sibility of the accused’s post-offence conduct on the issue of whether he intended to kill (or intended to cause bodily harm that he knew was likely to cause death to) his wife. I want to focus here on only one part of that evidence. The accused, a trained nurse, failed to perform CPR or to call 911 once he realized that his wife was unconscious following their violent confrontation. Instead, he wrapped her body up in the living room carpet and dragged it into the bedroom. More than three hours later he notified 911 of his wife’s death. Angelis testified that his failure to act was as a result of shock.27 Nevertheless, it is hard to imagine more compelling evidence of one’s purpose than the failure to do what you are trained to do — provide life- saving treatment in stressful and traumatic situations — to the mother of your children while they watched in horror. It becomes even more proba- tive when one considers the evidence of their daughter who testified that her mother had asked her to call 911 and get help during the confronta- tion, at which point the accused said “no, no, no” and then put his hand over his wife’s mouth.28 Moreover, the manner of killing in Angelis ar- guably involved more deliberate conduct than the pulling of a trigger, which occurred in White, which can happen accidentally. And to address
26Ibid. at para. 126. However, Justice Charron disagreed with Rothstein J. that there was sufficient evidence to conclude whether there was hesitation or not. 27See Angelis, supra note 6 at para. 28. 28As summarized at para. 7 of the respondent’s factum. See further, R. v. Ange- lis, 2010 ONSC 4312 (Ont. S.C.J.) at para. 18 (motion to quash the accused’s committal to stand trial) [Angelis 2010]; and R. v. Angelis, 2011 ONSC 462 (Ont. S.C.J.) (parole ineligibility sentencing decision) [Angelis 2011]. 346 CRIMINAL REPORTS 99 C.R. (6th)
Justice Binnie’s concern in White, failing to perform CPR or call 911 is not demeanour evidence and is more objective and discernible than evi- dence of a lack of hesitation. Thus, one would have thought that in applying White, the Court of Ap- peal would have concluded that the evidence in this case was probative of intent. But they disagreed. As a matter of precedent, Justice Laskin, for the Court, held: Even if one were to focus on what was likely the most cogent of the appellant’s post-offence conduct — the first category, namely, his failure to administer CPR to Lien or to immediately call 911 — I am not persuaded that his conduct could rationally support an inference of an intent to kill, rather than simply an inference of having done something wrong. Indeed, recent case law from this court suggests that an accused’s failure to render assistance after learning the victim may be dead is not probative of an accused’s level of culpability: see R. v. Anthony, 2007 ONCA 609, 228 O.A.C. 272, at paras. 52–58; R. v. Cudjoe, 2009 ONCA 543, 251 O.A.C. 163, at para. 88; R. v. McIn- tyre, 2012 ONCA 356, 291 O.A.C. 359, at para. 40.29 This reliance on prior jurisprudence is arguably problematic because it is referred to as if the Court has decided that as a general rule failing to render assistance can never be used as evidence of intent where the ac- cused admits the actus reus. This approach is inconsistent with White which makes it clear that there are no blanket rules to be applied with post-offence conduct. But the Court of Appeal’s analysis of the issue did not end with reliance on its prior jurisprudence. It engaged in inductive reasoning presumably in an effort to address why failing to render assistance in the context of this case was less probative and more speculative than the evidence in White. Justice Laskin held: . . . As a matter of logic and human experience, the appellant’s post- offence conduct could not support a rational inference of an intent to kill. That it could not do so is evident from the circumstances. The appellant and his wife had no history of violence or abuse in their relationship. Yet, they had just had a sudden and very physical alter- cation. The altercation occurred in front of their two children. It was brief. It left the appellant disoriented and bleeding profusely from his
29Angelis, supra note 6 at para. 58. Two of the cases cited by Justice Laskin (Anthony and Cudjoe) were cases of femicide. Angelis: Inductive Reasoning, Post-Offence Conduct and Intimate Femicide 347
genitals. And when it was over he knew only two things: Lien was dead, and he had killed her. In these circumstances, logic and human experience suggest that the appellant’s post-offence conduct was as consistent with a panicked reaction to Lien’s sudden and unintended death, as it was with a pan- icked reaction to her sudden and intended death. Thus, the jury should not have been repeatedly instructed that they could use this evidence to decide whether the appellant had the intent for murder.30 The critical question is whether the common sense and experience ap- plied by the court was reasonable. In exploring this question, one might ask where did the court get it from? It is certainly not clear. We do know that there is no reference to judicial experience or the relevant social sci- ence or academic literature. There is, in fact, social context evidence relevant to this question that could have been relied upon by the court. In 2010, Professor Isabel Grant published an article in which she reported the results of her study of re- ported intimate femicide cases in Canada from 1990 to 2008.31 She un- covered 252 Canadian cases with 216 of those involving appellate courts.32 These cases provide an objective record of the lived exper- iences of men and women in intimate relationships that result in death. In her review of the cases, Professor Grant found a number of common themes or indicators: . . . men who kill their spouses often do so out of jealousy, posses- siveness, or to prevent the spouse from leaving the relationship or entering a new relationship. . . . When men kill their intimate part- ners, the killing is sometimes the final act of violence against a spouse after a period of repeated abuse. . . . Some researchers iden- tify the underlying dynamic for men who kill their spouses or former spouses as men’s proprietary claim over women. The idea that “if I cannot have her, no one will” fits with the reality that women are most likely to be killed when they attempt to leave the relationship. . . . While each of these cases involved its own unique mix of brutal- ity and loss, after reading more than 250 judgments, one is struck by the similarity of the cases. The relationship has often been character- ized by ongoing violence or persistent arguments. The quarrels are
30Ibid. at paras. 56–57 (emphasis added). 31Grant, “Intimate Femicide”, supra note 1. 32Ibid. at 783–785. 348 CRIMINAL REPORTS 99 C.R. (6th)
frequently triggered by the accused’s suspicions of infidelity, which are occasionally substantiated, but more often not. Alcohol is very often a precipitating factor even where the state of intoxication is not sufficient to reduce murder to manslaughter. The victim is often also intoxicated. While there are some cases where the woman is killed by a single stab wound or gunshot, the degree of “overkill” in these cases is very disturbing. It is not uncommon to see descriptions of 20–40 stab wounds, or cases where multiple means are used to cause death, such as stabbing, strangulation, and a beating. Where there is evidence of provocation on the part of the deceased, it is often verbal, rather than physical, and trivial in nature.33 In writing this piece, I examined reported appellate cases from 2009–2012 and found similar indicators. These include: • a desire to maintain control, power and domination;34 • suspicions about an affair35 or other perceived challenge to the accused’s “manhood”;36 • gratuitous and sexualized violence;37 and
33Ibid. at 780–781, 785. 34See, for example, R. v. Mantley, 2013 NSCA 16 (N.S. C.A.) (husband at- tempted to kill his wife when she moved out of the family home). See further, R. v. R., 2012 BCCA 416 (B.C. C.A.); R. v. Flores, 2011 ONCA 155, 83 C.R. (6th) 93 (Ont. C.A.); R. v. Martel, 2011 ABCA 114 (Alta. C.A.); R. v. Jobateh, 2011 ONCA 171 (Ont. C.A.); R. v. G. (K.), 2010 ONCA 177 (Ont. C.A.); and R. v. Candir, 2009 ONCA 915 (Ont. C.A.). 35See, for example, R. v. Cudjoe, 2009 ONCA 543, 68 C.R. (6th) 86 (Ont. C.A.) [Cudjoe] (Cudjoe’s wife announced that she was leaving him to be with another woman). See further, R. v. Purdy, 2012 BCCA 272 (B.C. C.A.); and R. v. Kokotailo, 2011 BCCA 465 (B.C. C.A.). 36See, for example, R. v. Kimpe, 2010 ONCA 812 (Ont. C.A.) at para. 5 (ac- cording to Kimpe, his common-law spouse “taunted him about his poor sexual performance and declared that she was going to bring home another man who could satisfy her sexual needs. She suggested that the appellant could listen to them having sex”). 37In R. v. Panghali, 2012 BCCA 407 (B.C. C.A.), for example, the deceased, who was four months’ pregnant, was stabbed, burned and left on a beach. See further, R. v. Damin, 2012 BCCA 504 (B.C. C.A.) (126 stab wounds); R. v. Czibulka, 2011 ONCA 82 (Ont. C.A.) (the deceased “had been brutally beaten to death. Her eyes were swollen shut and bruises covered her whole body. Her ribs Angelis: Inductive Reasoning, Post-Offence Conduct and Intimate Femicide 349
• collateral damage with the woman’s new partner or others being killed or injured.38 Although the Court of Appeal in Angelis was prepared to take into ac- count context when engaging in inductive reasoning, it seems to have taken a narrow view of femicide by focussing on the absence of a history of physical abuse and in constructing the accused as a “mild mannered civil servant.” Given what we know about intimate femicide, what evi- dence did the court fail to properly consider or give sufficient weight to in assessing whether intent was a reasonable inference from the ac- cused’s failure to perform CPR or call 911? At the time of the killing, the relationship between Angelis and his wife had become “highly acrimonious.” As the trial judge observed, “[t]he of- fender had . . . recently become aware that his wife had a lover and was leaving him to live with her lover.”39 Indeed, two days before the killing, the accused told a co-worker that “his wife had been cheating on him for 15 years.”40 Angelis and his wife were also preparing for a “bitter cus- tody” battle. The deceased “believed that the appellant was trying to alienate her daughter from her and keeping both children away. He had accused her of being an unfit mother.”41 As noted in a pre-trial applica- tion ruling: Their domestic struggle had come to a point by then at which he ap- peared distraught and his life in shambles. Her circumstances were looking better and better. And now, he found himself in a position
were fractured in 25 places. Her diaphragm was lacerated and her breast bone was broken”); and R. v. Neumann, 2010 BCCA 446 (B.C. C.A.). 38See, for example, Cudjoe, supra note 35, where the accused’s wife and her new partner were both stabbed. See further, R. v. Tran, [2010] 3 S.C.R. 350, 80 C.R. (6th) 1 (S.C.C.) where the accused’s wife survived but her partner was killed. 39Angelis 2011, supra note 28 at para. 10. 40As summarized in Angelis 2010, supra note 28 at para. 24. 41Quoting from paragraph 5 of the respondent’s factum. It is troubling that none of this evidence was referred to in the decision. Rather, the court only relied on evidence to suggest that it was the deceased who was controlling and making life for the accused “unbearable.” 350 CRIMINAL REPORTS 99 C.R. (6th)
which might drastically reduce his chances of keeping the chil- dren — a possibility of the greatest dread to him.42 And finally, shortly before the killing, the accused was overheard calling his wife a “bitch.” All of this evidence suggested that while there may not have been a history of violence, a common feature of intimate femi- cide, there were many other indicators. Thinking about the killing of women by their intimate partners as a sys- temic problem with common characteristics is important and necessary in engaging in inductive reasoning because it helps rebut the commonly held belief that these killings are isolated or episodic; often the product of accident, provocation or intoxication; and where there is no evidence of prior violence, “out of character.” It would appear that this commonly held belief was applied by the Court of Appeal in this case. This social context evidence may also serve to reveal the unreasonableness of an inference the court is prepared to draw. So, for example, in this case, the Court of Appeal appears to have given considerable weight to the fact that the altercation took place in front of the accused’s children in sup- port of its conclusion on the post-offence conduct. However, as Professor Grant points out, “[w]omen are at greatest risk in their own homes, and the presence of children in the home does not appear to be a protective factor. For example, in a study of femicides from 1974–94, 100 children witnessed their mothers’ deaths.”43 It is hoped that in future cases more attention will be given to this social context evidence in intimate partner homicide cases and, more generally, to the process of inductive reasoning.
42Angelis 2010, supra note 28 at para. 32. 43Grant, “Intimate Femicide”, supra note 1 at 780–781. Indeed, in one of the leading Supreme Court of Canada cases, John Seymour stabbed his wife six times in front of their children. See R. v. Seymour, [1996] 2 S.C.R. 252, 49 C.R. (4th) 190 (S.C.C.).